RE [scilab-Users] termination indicator in lsqrsolve

[Michael BAUDIN]

Hi,

In the source code, we find the following comments :

info = 2  relative error between two consecutive iterates is at most xtol.
info = 5  number of calls to fcn with iflag = 1 has reached maxfev.

Notice that this is different from the content of the help page (by the 
way, this is a bug which should be reported). So I guess that the info=2 
either means that the solution is found, or that the algorithm stagnates. 
Did you check the value of the gradient and the eigenvalues of the Hessian 
matrix ? You can do this with the "derivative" function and the "order" 
option.

One thing that we may try in this situation is to compute the derivatives 
by different ways, with the goal of measuring the sensitivity of the 
solution to the accuracy of the gradient. Did you try to compute the 
gradient with "derivative" and various formula orders ? You can do this 
with the fjac argument of lsqrsolve.

Best regards,

Michael

PS
http://gitweb.scilab.org/?p=scilab.git;a=blob_plain;f=scilab/modules/optimization/src/fortran/minpack/lmder.f;hb=HEAD
http://gitweb.scilab.org/?p=scilab.git;a=blob_plain;f=scilab/modules/optimization/src/fortran/minpack/lmdif.f;hb=HEAD





ezequielsoule at gmail.com 
28/05/2012 19:13
Veuillez répondre à
users at lists.scilab.org


A
users at lists.scilab.org
cc

Objet
[scilab-Users] termination indicator in lsqrsolve






hello, I´m using the function lsqrsolve and I wonder what is the 
difference
between the termination indicator 2 and 5, and what can I do if I get the
"2".

2: number of calls to fcn reached
5: number of calls to fcn reached or exceeded maxfev

I´m getting the termination indicator 2, and there is no way to avoid it. 
I
increased maxev but nothing happens, so I can´t get a good solution.

Thank you,


--
View this message in context: 
http://mailinglists.scilab.org/termination-indicator-in-lsqrsolve-tp4019532.html

Sent from the Scilab users - Mailing Lists Archives mailing list archive 
at Nabble.com.

--
To unsubscribe from this mailing-list, please send an empty mail to
users-unsubscribe at lists.scilab.org
To check the archives of this mailing list, see
http://mailinglists.scilab.org/





Ce message et toutes les pièces jointes (ci-après le 'Message') sont établis à l'intention exclusive des destinataires et les informations qui y figurent sont strictement confidentielles. Toute utilisation de ce Message non conforme à sa destination, toute diffusion ou toute publication totale ou partielle, est interdite sauf autorisation expresse.

Si vous n'êtes pas le destinataire de ce Message, il vous est interdit de le copier, de le faire suivre, de le divulguer ou d'en utiliser tout ou partie. Si vous avez reçu ce Message par erreur, merci de le supprimer de votre système, ainsi que toutes ses copies, et de n'en garder aucune trace sur quelque support que ce soit. Nous vous remercions également d'en avertir immédiatement l'expéditeur par retour du message.

Il est impossible de garantir que les communications par messagerie électronique arrivent en temps utile, sont sécurisées ou dénuées de toute erreur ou virus.
____________________________________________________

This message and any attachments (the 'Message') are intended solely for the addressees. The information contained in this Message is confidential. Any use of information contained in this Message not in accord with its purpose, any dissemination or disclosure, either whole or partial, is prohibited except formal approval.

If you are not the addressee, you may not copy, forward, disclose or use any part of it. If you have received this message in error, please delete it and all copies from your system and notify the sender immediately by return message.

E-mail communication cannot be guaranteed to be timely secure, error or virus-free.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/users/attachments/20120529/51bd3942/attachment.htm>